Multiagent Systems
EAIR: Interaction Risks in Enterprise LLM Multi-Agent Systems
Yuchen Ma, Swadheen Shukla, Paulo Salem et al. (+5 more)
LLM
agents increasingly divide enterprise work across roles with different goals, authority, access, and information. Because each
agent often sees only a local snapshot of the enterprise workflow, locally reasonable actions can still violate organization-wide constraints (e.g., several service
agents can each issue credit that is individually permitted while collectively exceeding a shared compensation cap). We introduce EAIR, the first framework for analyzing enterprise multi-
agent interaction risks. We make four contributions. (1) We formalize enterprise interaction risk in terms of the organizational properties that workflow operations must preserve, and operationalize this formulation as a taxonomy of testable risks with explicit failure criteria and evidence requirements. (2) We build EAIR-Worlds, a suite of executable micro-worlds that instantiates selected risks as controlled, observable workflows with explicit roles, policies, permissions, source facts, and execution traces. (3) We develop a novel diagnosis-and-repair pipeline that uses the resulting execution traces to locate contributing interaction boundaries, guide targeted workflow repairs, and retest the repaired system. (4) Large-scale experiments show recurring interaction failures across the organizational properties studied. Together, EAIR is the first operational framework for diagnosing and mitigating enterprise multi-
agent interaction failures.
Read the full abstract ▾
All authors Yuchen Ma, Swadheen Shukla, Paulo Salem, Yue Huang, Robert Sim, Stefan Feuerriegel, Michel Galley, Jianfeng Gao
Artificial Intelligence
CESNET-TimeSeries26: A Multi-Year Network Traffic Time Series Dataset for Annual Seasonality, Concept Drift, and Long-Term Anomaly Detection
Josef Koumar, Karel Hynek, Kamil Jerabek et al. (+1 more)
Forecast-based models are fundamental to network traffic management and anomaly detection. However, it is essential to evaluate these forecasting models over extended periods of time in order to assess their adaptability to unexpected phenomena and shifts in underlying patterns. This kind of long-term evaluation requires multi-year data to account for yearly seasonality and behavioral drift. Existing public
datasets are too short to support such long-horizon forecasting tasks. We introduce CESNET-TimeSeries26, comprising 995 days of continuous traffic time series from the CESNET3 network, together with a single subnet spanning approximately 4.5 years. The
dataset captures 278,466 IP addresses, 285 institutions, and 566 institutional subnets at three aggregation intervals. The
dataset provides a definitive longitudinal
benchmark for evaluating forecasting models in both operational management and security monitoring. We demonstrate that annual seasonality is the dominant periodic component. Moreover, we show that a forecasting model can exploit the annual component only when its training window covers at least one complete annual cycle, which no sub-year
dataset provides.
Read the full abstract ▾
All authors Josef Koumar, Karel Hynek, Kamil Jerabek, Pavel Siska
Distributed Computing
Accelerating MoE Training through Expert Placement and Replica Coordination
Zhiqiang Tan, Maoxin Wang, Wenxiang Lin et al. (+4 more)
Distributed training of sparsely activated mixture-of-experts (MoE) models is often bottlenecked by workload imbalance and all-to-all (A2A) communication induced by dynamic
token routing. Existing
token deduplication, expert placement, and replication techniques alleviate these bottlenecks but do not jointly capture their distinct load semantics. Under hierarchical
token deduplication, communication counts each
token once per destination group at each hierarchy level, whereas expert computation processes every
token-expert assignment. With replication, both costs depend jointly on physical layout and
token-to-copy mapping, making separate optimization suboptimal. We present PlaceMoE to address this coupling. First, we formulate the joint optimization of expert layout and source-aware
token-to-copy mapping for E experts distributed across P devices, accounting for both hierarchical communication traffic and expert-compute load. Second, we develop a solver that combines co-selection-aware expert grouping, bounded dynamic programming for replica allocation, and alternating refinement of layout and mapping. Third, PlaceMoE overlaps replica-gradient synchronization with
attention backward. We evaluate PlaceMoE on a 32-GPU cluster and a 64-NPU cluster using Qwen3-VL and DeepSeek-V3. Experimental results show that PlaceMoE achieves up to a 6.94x speedup in A2A communication and 1.74-2.33x speedups in end-to-end training over the state-of-the-art training framework VeOmni.
Read the full abstract ▾
All authors Zhiqiang Tan, Maoxin Wang, Wenxiang Lin, Sijie Wang, Qiang Wang, Shaohuai Shi, Bo Li
Cryptography and Security
Protecting Shared Agent Workflows from Invalidation Attacks with Signed Trajectories and Runtime Verification
Ruixiao Li, Fahao Chen and Peng Li
Large
language model (LLM)-based
agents increasingly rely on shared workflow repositories to transfer experience across tasks and
agents. By abstracting past executions into reusable workflow graphs, such systems significantly improve planning efficiency and generalization. However, this emerging paradigm introduces a new and underexplored security risk: if the shared repository is compromised, tampered workflows can systematically mislead downstream
agents, inducing persistent execution failures or unsafe behaviors. In this work, we present a principled protection framework for shared
agent workflows based on signed trajectories and runtime verification. We require all cloud-collected trajectories used for workflow generation to be cryptographically signed, with each trajectory representing a concrete execution instance. Prior to repository inclusion, trajectories undergo a unified screening process to ensure correctness and validity. When an
agent retrieves a workflow from the shared repository, it also obtains a small set of signed supporting trajectories and verifies both their cryptographic integrity and their semantic consistency with the workflow. To make this practical for branching workflows, we further propose a runtime, path-aware
retrieval strategy that dynamically fetches only the signed trajectories relevant to the execution path actually taken by the
agent, substantially reducing communication and verification overhead. Extensive experiments across multiple
reasoning benchmarks show that workflow invalidation attacks achieve up to 90.12% success rates, degrading
agent inference performance by up to 73.16 percentage points in accuracy without effective workflow protection. In addition, the proposed protection achieves an average F1 score of 85.26% and a detection rate of 79.97%, outperforming existing baselines by about 2.02x.
Read the full abstract ▾
Networking
Energy-Aware Agentic Orchestration for Multi-Site Service Deployment
Rodrigo Moreira, Alisson Chaves, Larissa F. Rodrigues Moreira et al. (+6 more)
Energy-aware orchestration across geographically distributed infrastructures is increasingly important as data centers absorb growing Artificial Intelligence (AI) workloads. However, existing approaches often exclude the energy cost of automated decision-making from system-level sustainability assessments. This paper presents an energy-aware agentic orchestration framework that treats Large
Language Model (LLM)
reasoning as a measurable component of the control loop. We compare single-shot tool calling, ReAct, and plan-and-execute strategies on a multi-site Kubernetes testbed spanning two European data centers and instrumented at layers ranging from power distribution units to per-pod energy attribution. Using ten natural-
language orchestration intents with varying decision complexity, we show that
reasoning strategy and intent complexity strongly influence graphics processing unit (GPU) energy consumption during orchestration decisions. ReAct and plan-and-execute generated executable tool calls across all evaluated intents but required multiple LLM inferences, whereas the single-shot configuration was less costly but non-actionable. Prefix caching reduced mean ReAct GPU energy consumption by 29.4%. For the evaluated lightweight workload, the infrastructure-energy changes observed within the one-second measurement window were insufficient to offset the measured GPU decision energy. These findings demonstrate the importance of accounting for
reasoning overhead when assessing the sustainability of AI-native orchestration and motivate energy-budgeted selection of
agent strategies.
Read the full abstract ▾
All authors Rodrigo Moreira, Alisson Chaves, Larissa F. Rodrigues Moreira, Tereza Carvalho, Tiago Barros, Daniel Corujo, Rui Aguiar, David Gomez-Barquero, Flávio de Oliveira Silva
Artificial Intelligence
GREAT-ESDD: Graph Edge Attention Network for Effectively Steered Delivery Decisions
Attila Lischka, Ze Zhou, Xuegang Ban et al. (+1 more)
Online, same-day delivery requires operators to constantly update their routes, and the routing cost for deliveries increases with the number of customers. Out-of-home (OOH) delivery options can help companies handle increasing customer numbers while keeping delivery costs low. We develop a dynamic pricing-and-routing framework for this online ordering scenario with home and OOH delivery options. First, we develop a neural solver for the capacitated vehicle routing problem (CVRP) based on the Graph-Edge-
Attention-Network (GREAT). We train it to solve CVRP with varying instance sizes and starting conditions, allowing it to reoptimize partially executed routes under changing conditions. It is fast while achieving solution quality competitive with traditional baselines. Our solver processes several CVRP instances in parallel in batches and we use it to solve the candidate routing problems that arise from an incoming customer's choice of home or OOH delivery. This provides an estimate of the delivery cost increase per option, which we use to set dynamic prices. Further, it directly provides the updated route once the customer decides. We trade off delivery fees against routing costs and steer customers toward OOH delivery when home delivery would be costly. However, effective steering requires us to know the customer preferences. To learn those online, we use Thompson sampling with a Laplace-approximated posterior. We call the resulting framework GREAT-ESDD (GREAT for Effectively Steered Delivery Decisions). Compared to a traditional routing solver, it increases profit by around 11% under a similar pricing decision time budget. Moreover, we show that we can estimate customer preferences successfully, outperforming a static pricing strategy after only 5 days of simulated operations.
Read the full abstract ▾
All authors Attila Lischka, Ze Zhou, Xuegang Ban, Balázs Kulcsár
Artificial Intelligence
Counterfactual Regret for Reflective Agents with Lean-Verified Guarantees
Zhongnian Li, Shi Xu, Mengting Xu et al. (+2 more)
Reflective
language agents reuse feedback and
memory to adapt across tasks without updating model weights. However, existing guarantees for policy selection and action generation do not by themselves establish when local reflection yields long-term performance guarantees for the complete
agent under resource constraints. In this paper, we address this gap by deriving sufficient conditions for sublinear counterfactual regret from local generation and deployment guarantees, while retaining the native controller. Specifically, we establish a resource-sensitive transfer that bounds full-state deployment discrepancies along actual execution paths and preserves the own-history laws of both learner and comparator. Besides, local generation reliability yields a cumulative generation opportunity bound, linking policy acquisition to long-term performance through an explicit quality, coverage, and hard-budget tradeoff. To verify these guarantees, we formalize the conditional proof chain for all thirteen numbered theoretical results in Lean, including a native instance with repeated policy growth and imperfect validation.
Read the full abstract ▾
All authors Zhongnian Li, Shi Xu, Mengting Xu, Peng Ying, Xinzheng Xu
Computation and Language
Training-Free BioNEL-RAG: Advancing Biomedical Entity Linking through Retrieval-Augmented Generation and Large Language Model
Ziyi Chen, Hui Liu, Ziyang Ling et al. (+2 more)
Biomedical Named Entity Linking (BioNEL) is the task of mapping mentions extracted from biomedical texts to canonical entities or concepts in the standardized knowledge base. Although existing methods have achieved remarkable performance in mapping accuracy, most rely on training with domain-specific
datasets, which fundamentally limits their cross-domain adaptability. To address this limitation, we propose BioNEL-RAG, a novel training-free framework that combines
retrieval-augmented generation (RAG) with large
language models (LLMs). Within this framework, the RAG module focuses on retrieving candidate entities, while the large
language model module plays the role of selecting the most appropriate candidate for entity normalization, which is based on the analysis of the mention's context, with the aid of its remarkable text comprehension capabilities. Experimental results demonstrate that BioNEL-RAG achieves performance comparable to existing methods without requiring training on subdomain
datasets, whereas these methods necessitate fine-tuning on specific subdomains such as genes, diseases, or chemical compounds.
Read the full abstract ▾
All authors Ziyi Chen, Hui Liu, Ziyang Ling, Ronald X.Xu, Mingzhai Sun
Machine Learning
Formalizing Forgetting: Theory of Forgetting for Online Continual Learning
Renjie Cao, Yifei Cheng and Li Shen
Online continual learning (OCL) requires a learner to acquire a sequence of tasks in a single pass over the data, and catastrophic forgetting of earlier tasks remains its central challenge. Despite the many methods proposed to mitigate forgetting, the theoretical understanding of forgetting in OCL is still limited: existing analyses mostly focus on linear, overparameterized, or neural tangent kernel regimes, and it remains unclear which factors govern forgetting and how they affect it. To fill this gap, we analyze single-pass Shuffle SGD under strongly convex and general convex objectives and derive explicit upper bounds on the expected final forgetting. Further analysis of these bounds shows how the per-task data amount, task similarity, and task ordering enter them, and that their effects can differ across objective classes. Moreover, experiments with neural networks on Split-CIFAR-100 and PMNIST show that the trends suggested by our theory also appear in practice, indicating that these insights extend beyond the convex setting.
Read the full abstract ▾
Systems and Control
Decision-Focused Learning for Optimal Battery Scheduling Under Load and Photovoltaic Power Uncertainty in Microgrids
Marios Saleptsis, Antoine Lesage-Landry, Marco Mussetta et al. (+1 more)
The increasing penetration of renewable energy sources in microgrids demands accurate day-ahead forecasting to support optimal battery energy storage system (BESS) scheduling. Yet the conventional predict-then-optimize paradigm trains forecasters on statistical accuracy metrics that are agnostic to the asymmetric and time-varying cost impacts of forecast errors. This work proposes a decision-focused learning (DFL) framework that trains gated recurrent unit forecasters for both load demand and photovoltaic generation directly on the downstream operational cost of a two-stage energy management system, where day-ahead BESS commitments are fixed prior to real-time balancing. Gradients are backpropagated from the real-time operational cost through both optimization stages to the forecaster weights via implicit differentiation, enabling the forecasters to internalize the penalty structure of the scheduling problem. The framework is validated on the MG2Lab experimental microgrid testbed at Politecnico di Milano across two operational scenarios of increasing grid constraint and eight forecasting configurations comprising single-source and combined-forecaster settings. Results illustrate that the DFL-trained pipeline reduces operational cost by 3.78% under normal grid conditions and by 30.60% under grid-constrained conditions relative to the accuracy- optimized baseline, while nearly eliminating load shedding events. Notably, the DFL forecasters exhibit higher statistical errors than their accuracy-trained counterparts yet achieve consistently lower operational costs, confirming that forecast accuracy and decision quality do not share a monotonic relationship in constrained microgrid scheduling.
Read the full abstract ▾
All authors Marios Saleptsis, Antoine Lesage-Landry, Marco Mussetta, Sonia Leva
Logic in Computer Science
StructHCP: Structure-Aware SAT–CEGAR for the Hamiltonian Cycle Problem
Quan Van Vi, Tuyen Van Kieu and Khanh Van To
Many exact
Hamiltonian-cycle solvers use the same edge-level representation for every input, even when graph structure admits a smaller set of global decisions. We present StructHCP, a SAT–CEGAR pipeline that selects a representation through inexpensive structural recognizers. Specialized searches handle high-degree clusters and paired chains, while complete directed and undirected encodings handle cubic and general graphs. Every returned cycle is reconstructed when necessary and verified in the input graph. On the 1,001-instance FHCP Challenge Set, StructHCP solves 826 instances, compared with 741 when the same general SAT–CEGAR engine is applied directly to every unreduced input graph. Relative to this ablation, StructHCP reduces PAR-2 by 32.9 percent. Formulation selection solves 82 more instances than the contracted general encoding; high-degree-cluster and cubic graphs account for 81 of this difference. StructHCP also solves 34 of the 51 instances solved by neither HCup nor Cut-set CEGAR, raising the combined coverage of the three solvers from 950 to 984. Twelve of its solutions lie outside the union of published per-instance SAT results analyzed here, and six also lie outside the published exact result sets in our comparison. These results show that selecting the representation can complement connectivity refinement on structured HCP instances.
Read the full abstract ▾
Artificial Intelligence
LYNX: Analytical Program Synthesis via Grounding and Closed-Loop Refinement
Ruichen Xia, Xiao Han, Dingyu Yang et al. (+5 more)
LLM-powered data analytics aims to translate natural
language instructions into executable analytical programs over underlying data. However, existing approaches, whether treating LLMs as intent-to-code translators or as standalone analytical engines, often fail to reliably align domain semantics with complex schemas and heterogeneous metadata, resulting in brittle execution and semantically inconsistent analytical outputs in real-world workloads. We present LYNX, a multi-
agent framework that formulates LLM-driven analytics as a schema- and knowledge-constrained analysis problem rather than a single-pass generation task. LYNX organizes analytical synthesis into a closed-loop workflow of planning, generation, verification, and refinement through three role-specialized
agents that collaboratively enforce schema grounding, semantic consistency, and execution correctness. To address domain ambiguity and schema misalignment, LYNX integrates Semantic-Structural
Distillation and Stage-Aware Allocation (SSD-SAA). A delta-guided iterative revision continuously diagnoses and repairs semantic and execution-level failures, substantially improving synthesis robustness. We evaluate LYNX on two NL-to-SQL
benchmarks, BIRD-SQL Mini-Dev and the more challenging Spider 2.0-Lite. Experimental results demonstrate strong performance and efficiency, achieving 10.66% to 59.54% execution accuracy across lightweight and commercial models, while reducing
token consumption by 70.53% compared with the prior most efficient method. Our framework displays high generalization over different backbones, from open-source, lightweight models to the latest commercial models. Additional case studies on quantitative finance analytics and complex multi-relational SQL workloads further demonstrate the framework's robustness, scalability, and practical applicability in real-world analytical scenarios.
Read the full abstract ▾
All authors Ruichen Xia, Xiao Han, Dingyu Yang, Luwei Yang, Sai Wu, Dongxiang Zhang, Binhui Yao, Gang Chen
Sound
Training a Compact, Strong, and Realistic Text-to-Speech Model
Runyuan Cai, Yu Lin, Yiming Wang et al. (+1 more)
Text-to-
speech (TTS) systems are conventionally optimized for intelligibility and speaker similarity. However, neither objective captures the prosodic detail that distinguishes human performances from synthetic
speech, leaving a realism gap that larger models and higher-quality data can narrow but not close. In the absence of an objective that directly targets this gap, realism remains a persistent challenge in TTS. We present Audio8-TTS, a compact 0.6B-parameter model for zero-shot voice cloning that achieves state-of-the-art performance in intelligibility and speaker similarity, with particularly strong results in perceptual naturalness. To align the training signal with human perception rather than signal-level fidelity, we construct a quality ladder from widely used open-source TTS systems and train a realism critic to assess generated
speech against matched human recordings of the same sentences. An identity-erasure objective discourages the critic from relying on generator fingerprints. We train the model in three stages: multilingual supervised fine-tuning; group relative policy optimization (GRPO) with word error rate (WER) and speaker similarity (SIM) rewards; and GRPO guided by the realism critic. The final stage improves perceived prosodic naturalness and speaker similarity while preserving WER performance.
Read the full abstract ▾
All authors Runyuan Cai, Yu Lin, Yiming Wang, Xiaodong Zeng
Artificial Intelligence
ProgAgent: Learning Tool Orchestration through Programmatic Training
Hao Li, Keqing He, Longtao Zheng et al. (+4 more)
Large
language models (LLMs) are increasingly expected to orchestrate diverse tools over long-horizon tasks. Existing
agents typically use Direct Tool Calling (DTC), where the model invokes tools sequentially, and tool responses accumulate in the model context, which can lead to context rot and orchestration breakdown. In contrast, Programmatic Tool Calling (PTC) invokes tools through executable code and stores intermediate results in runtime variables rather than appending them to the context. Prior work has explored PTC mainly as an inference-time optimization, but its benefits for agentic training remain unclear, and suitable verifiable training tasks are scarce. In this work, we introduce ProgAgent, a full PTC training stack, comprising a pipeline for synthesizing verifiable long-horizon tasks, a lightweight PTC runtime, and a two-stage training recipe. Using only 215 tasks instantiated at four environment scales, ProgAgent yields higher pass rates than the evaluated baselines of comparable size across four
benchmarks. On LOCA-bench, ProgAgent-14B achieves a 30.7% pass rate at a 64k-
token environment state, where these baselines score zero. Cross-harness experiments show that PTC-trained models outperform DTC-trained models under both evaluation interfaces, suggesting that the benefits of PTC training extend beyond the interface.
Read the full abstract ▾
All authors Hao Li, Keqing He, Longtao Zheng, Junlong Li, Weihao Zeng, Qian Liu, Yuzhen Huang
Artificial Intelligence
How Should a Camera-First AI Agent Work?
Xiaofan Wu, Xi Zeng, Miaoxia Chen et al. (+8 more)
Camera-first
agents must act on the visual world as it unfolds. We argue that the central obstacle for future camera-first
agents is not recognition ability -
multimodal models (
vision-
language and live/streaming alike) generalize well through
language and remain the right tool for semantic
reasoning - but time. Many visual targets are ephemeral: they appear, are query-relevant, and leave within a short window. Binding
memory to the same end-to-end path as
reasoning means that by the time a monolithic pipeline completes, the evidence is gone. We formalize this as Ephemeral-Target Interaction (ETI) and a Perceptual Deadline Contract separating a hard
memory deadline from a soft
reasoning deadline, grounded in the bounded, passive nature of sensory and working
memory. We propose a two-deadline architecture: an always-on, bounded, non-semantic fast path; an active remembering mechanism that commits a moment to durable
memory before it is overwritten (event-driven in v1); and a deferred deliberation layer (
multimodal model + tools). We instantiate the contract on real MOT17 track windows, a consistency check, and calibrate the single-path baseline from measured frontier-model latency: even a short answer takes 1.2-9.3 s, and an agentic identify-and-search longer. Answers can be late; evidence cannot be missed.
Read the full abstract ▾
All authors Xiaofan Wu, Xi Zeng, Miaoxia Chen, Peishan Chen, Shuyan Li, Chen Wang, Xiaohan Wang, Wanling Xu, Jiyun Yao, Hanyong Zhong, Jiahao Zhu
Machine Learning
Prior-Fitted and Frozen Scorers for Multimodal Contrastive Learning
Tillmann Rheude, Roland Eils and Benjamin Wild
Multimodal contrastive learning typically relies on manually specified scoring functions, such as cosine similarity, to quantify the similarity between representations from different modalities. Several such functions have been recently proposed to extend contrastive learning beyond two modalities. In principle, however, any function that assigns high compatibility to matched samples and low compatibility to mismatched ones could serve as a scorer. Motivated by this broader view, we instantiate the scorer with a
Transformer and find that even a randomly initialized model, when kept frozen, can outperform strong handcrafted
multimodal contrastive baselines, whereas optimizing the scorer jointly with the encoders is less effective. Building on this observation, we introduce the Prior-Fitted Scorer (PFS): a
Transformer scorer pre-trained on synthetic
multimodal data generated from structural causal model (SCM) priors before transferring it to real-world data. The SCM exposes the scorer to diverse dependency structures, including shared, pairwise, and private information, providing a structured prior over higher-order cross-modal similarity. Across four real-world
datasets, frozen
Transformer scorers are competitive to contrastive baselines, and prior-fitting further improves the random initialization. More broadly, our results suggest that
multimodal contrastive learning need not rely on manual similarity functions:
Transformers can provide strong inductive biases, while causal pretraining can further shape these biases.
Read the full abstract ▾
Computation and Language
Who Receives Respect? A Counterfactual Benchmark for Honorific Allocation Bias in Bengali
Ajwad Abrar, Md Kamrul Hasan and Hasan Mahmud
Bengali has no grammatical gender, but honorificity is obligatory: every third-person reference must choose ordinary or respectful register. English has no such signal, so the system must choose a level of deference that the source never specified. We use this property to measure how large
language models (LLMs) allocate respect in English–Bengali translation. Our evaluation set contains 2{,}016 sentences across 72 occupations. Each appears in seven sentence frames under four pronoun conditions. Masculine and feminine versions differ by one word, forming counterfactual pairs. We report a single Honorific Bias Score (HBS) for each model on a 0–100 scale, covering both the gender and occupation axes. A lower score means that respect is allocated more evenly. For the occupation analysis, we compute each occupation's honorific rate across the seven no-pronoun sentence frames. The standard deviation of these 72 occupation-level rates ranges from 17.2 to 35.5 percentage points across models, showing substantial variation in how honorifics are allocated across occupations. Occupation is associated with 3.2% to 51.5% of the variance in register choice, compared with at most 1.2% for gender. Gender effects are much smaller overall: only two of the nine models show significant directional gender gaps, and they favor opposite genders. HBS ranges from 19.9 to 70.1 across the nine models, showing substantial variation in how evenly the models allocate respect.
Read the full abstract ▾
Software Engineering
Do Modifications Made by Coding Agents Make Subsequent Software Maintenance Harder?
Ruwei Pan, John Zhang, Linhao Wu et al. (+3 more)
Coding
agents increasingly maintain repositories over time, where each change becomes the starting point for subsequent tasks. Yet existing
benchmarks either start tasks from developer-written code or cannot separate later-task difficulty from the effects of an
agent's earlier changes. We ask whether
agent-written code makes subsequent maintenance harder and whether the cost stems from inherited test failures or the code itself. We introduce SWE-Maintain, comprising 100 three-event maintenance episodes from 50 real-world Python repositories, with hidden tests certified across each sequence. Each later task is evaluated twice with the same
agent and tests: from the developers' preceding patches (Developer-History) and from the
agent's own changes (
Agent-History). Across five coding systems, two backbone models, and 1,996 paired later events,
Agent-History reduces later-task resolution by 5.6 percentage points on average. When success additionally requires all earlier tests to pass, the gap reaches 14.0 points. Two-thirds of
Agent-History starting repositories already fail earlier tests, and only 1.9% of events starting from those states end with all accumulated tests passing. When the inherited code passes earlier tests and the next task is not already solved, the disadvantage disappears (+2.7 points; 95% CI [-0.5, +6.1]). These findings suggest that the main downstream cost of
agent-written changes lies in inherited test failures rather than a measurable maintenance disadvantage in code that preserves earlier tested behavior. Evaluations should therefore include
agent-produced history, and coding
agents should preserve earlier behavior across changes.
Read the full abstract ▾
All authors Ruwei Pan, John Zhang, Linhao Wu, Yakun Zhang, Hongyu Zhang, Lu Zhang
Cryptography and Security
When Does Trust Help? A Leakage-Free Evaluation of Representation- and Aggregation-Based Defenses Against Backdoors in Federated Intrusion Detection
Shiddarth Dey Tusar and Zawad Yalmie Sazid
Federated learning (FL) lets distributed intrusion-detection systems (IDS) train on private traffic, but exposes the global model to poisoning and backdoor attacks under non-IID client data. A popular defense family scores or filters clients using a small trusted "root"
dataset and the similarity of client behaviour to a trusted reference. I re-examine these trust-based defenses under a protocol designed to remove common evaluation artifacts: train-only feature standardization, a server root set disjoint from client data, multi-seed (n=6) runs with paired significance tests, formal (varepsilon, delta) accounting for DP-SGD, and a defense-aware adaptive attacker. Across three NIDS
datasets (CSE-CIC-IDS2018, NF-ToN-IoT, NF-UNSW-NB15-v2) I find: (1) a representation-similarity aggregator (RepGuard) significantly improves classification utility and seed-stability over FLTrust under heavy non-IID poisoning, and - under severe heterogeneity or many adversaries - also lowers backdoor ASR substantially (e.g., 0.48 vs FLTrust 0.86 at Dirichlet alpha=0.1); (2) at strong trigger magnitudes no defense reduces backdoor attack success rate (ASR approx 0.95); (3) a controlled trigger-magnitude sweep shows defenses help only for small triggers (ASR 0.10 for RepGuard at magnitude 0.5, vs. 0.40 undefended) and collapse to ASR approx 1.0 by magnitude 2; (4) a triggered-canary check that assumes knowledge of the attacker's trigger - the assumption behind much prior "success" - lowers ASR from 0.49 to 0.12, i.e., trigger knowledge, not representation scoring, suppresses the backdoor; (5) correctly-accounted DP-SGD (varepsilon in [0.3, 16]) costs little utility but does not mitigate the backdoor. I release the protocol so future FL-IDS defenses can be evaluated without the leakage and oracle assumptions that inflate reported robustness.
Read the full abstract ▾
Cryptography and Security
Last Secure Code Benchmark: Is the Code You Vibe Secure?
Tianyi Wu, Chao Ge, Peng Chen et al. (+12 more)
AI coding
agents now write much of the code that ships, yet whether that code is secure remains unclear. We introduce the Last Secure Code
Benchmark, a comprehensive evaluation of secure code generation by coding
agents. Built from 150 real vulnerabilities, its 450 tasks span seven widely used
languages (Java, C, C++, JavaScript, Python, PHP, and Go), 24 CWEs including 21 of the 2025 CWE Top 25 and 8 of the Top 10 KEV weaknesses, more than any
benchmark we compare with, and three vibe-coding scenarios: filling in a function, writing a file, and building a repository's source tree from scratch. Every task asks only for functionality, through a specification that never mentions security, and hidden functional and security tests grade the result. Across six frontier
agents, 73.6% to 91.3% of tasks pass functionally, yet even the strongest, Claude Fable 5, produces code that is both functional and secure for only 25.1% of tasks, and the others for 16.4% to 24.7%. The best model only passes 12% of the vulnerability scenarios across all vibe coding setting, posing strong concern on the security of the code written by vibe coding.
Read the full abstract ▾
All authors Tianyi Wu, Chao Ge, Peng Chen, Fengji Zhang, Yongxing Wu, Yuchong Xie, Jian Li, Chaoyuan Peng, Kaikai Zhang, Dong Huang, Mingzhe Du, Yukai Huang, Jiaheng Zhang, Terry Yue Zhuo, See-Kiong Ng
Information Theory
Sum Rate Maximization for Practical RIS-Enhanced Rate-Splitting Multiple Access Systems
Qian Zhang, Xuejun Cheng, Jiaming Shi et al. (+3 more)
Reconfigurable intelligent surfaces (RIS) combined with rate-splitting multiple access (RSMA) provide a promising solution for improving the spectral efficiency (SE) of sixth-generation (6G) communication systems. However, most existing studies neglect the practical coupling between the RIS phase shift and amplitude, which may cause phase-shift mismatch and significant performance loss. In this paper, we investigate the multiuser sum-rate maximization problem for practical RIS-enhanced RSMA systems considering amplitude-phase coupling. To address the resulting non-convex problem, an effective variable separation
algorithm is proposed. The proposed
algorithm is compatible with RIS reflecting optimization under the continuous phase shift or any amplitude-phase coupling relationship and can also be directly extended to practical RIS-enhanced NOMA and OMA systems. Simulation results demonstrate performance gain of over 190% compared with the ideal-RIS-based
benchmark, while the proposed RSMA scheme significantly outperforms the corresponding NOMA and OMA schemes.
Read the full abstract ▾
All authors Qian Zhang, Xuejun Cheng, Jiaming Shi, Yufei Zhao, Rugui Yao, Ye Fan
Numerical Analysis · Cross-listed from math.OC
A Unified Framework for Randomized Row-Action Methods for Structured Data Fitting and Constraint Geometry
Valentin Leplat, Zeyu Dong and Junfeng Yin
We study randomized row-action methods for problems with linear observations, convex constraints, and non-Euclidean geometry. Our starting point is a unified formulation in which the data-fitting term acts in observation space, while the feasible set and the regularizer encode structure in variable space. It covers constrained least squares, nonnegative and simplex-constrained models, regularized data fitting, and equality-constrained Bregman selection problems. The structure of these problems leads naturally to two algorithmic branches. The first uses primal Bregman-proximal row steps for soft data-fitting problems. The second uses dual Bregman–Kaczmarz steps for problems with hard equality constraints. We first derive a common one-step inequality. We then give deterministic and
stochastic convergence results for the soft branch under relative smoothness. For constrained least squares, the normalized row oracle has exact first- and second-moment identities. They yield an expected ergodic bound in the inconsistent case, an $O(1/K)$ bound in the interpolating case, and linear convergence of the actual iterates under a residual error bound. For equality-constrained selectors, we prove linear convergence in expected Bregman distance under a residual-to-Bregman error bound. We recall global sufficient conditions for this bound and derive local constants for the nonnegative quadratic geometry and the
entropy geometry on the simplex. Numerical experiments illustrate the noisy stepsize trade-off and the local
entropy selector theory, and show that entropic row-action updates can recover interpretable factors in a simplex-structured nonnegative
matrix factorization problem.
Read the full abstract ▾
Artificial Intelligence
CUSE: Fine-Grained Multimodal Unlearning with Counterfactual Guidance and On-Policy Self-Distillation
Chenlu Ding, Jiancan Wu, Yanchen Luo et al. (+4 more)
Multimodal large
language models (MLLMs) can reproduce sensitive, copyrighted, or unsafe information from their training data. Effective unlearning must remove target information while preserving non-target content, even within the same response. Existing methods typically apply uniform forgetting objectives to complete responses and rely on fixed, off-policy supervision, resulting in coarse
token control and misalignment with the model's current behavior. We propose CUSE, a Counterfactual-guided Unlearning framework via on-policy Self-
distillation for
multimodal large
language models. It first constructs counterfactual inputs by removing target-related evidence and uses the resulting changes in
token probabilities to identify which
tokens require stronger forgetting. It then distills
token-level forgetting and retention targets on responses sampled from the current policy, with a frozen copy of the original model serving as the teacher model augmented with the corresponding forgetting- and retention-oriented privileged contexts. This design jointly determines what to forget and what to preserve. On MLLMU-Bench, CUSE reduces forget-set ROUGE-L by 33.6% while achieving 1.2% higher retain-set ROUGE-L than the strongest baseline.
Read the full abstract ▾
All authors Chenlu Ding, Jiancan Wu, Yanchen Luo, Tianyu Zhang, Yancheng Yuan, Xiang Wang, Xiangnan He
Machine Learning
CrossMean: Model Merging by Pairing Merged and Expert Activations
Moshe Eliasof
Model merging combines fine-tuned experts into one network without retraining, and closed-form methods compute each merged layer from statistics of its input activations. However, these statistics are measured on a single expert, while in the merged model, each layer receives activations from the merged layers that precede it. Recent work re-measures these statistics in the merged model but still forms them from a single network. We formulate closed-form methods as a linear system and show that they use the same
matrix on both sides. We then introduce CrossMean, which replaces this
matrix on one side with a bilinear statistic that pairs merged and expert activations. From the same statistics, we also derive a label-free criterion that indicates the task sets in which accounting for the merged activations improves the merge more. Empirically, task sets with a lower criterion value show a larger improvement. On the eight-task
benchmark of each
vision backbone that we tested and on BERT encoders, CrossMean improves on merging with statistics measured on the experts, and with all linear layers merged, it also improves on re-measuring these statistics in the merged model. These results suggest that merging benefits from pairing merged and expert activations.
Read the full abstract ▾
Computation and Language
The Constitution Speaks Nheengatu: An Open MT System and Reproducible Corpus Pipeline for the Amazonian Língua Geral
Fabrício Carraro, Rodolfo Joel Zevallos, Rodrigo Gonçalves de Souza et al. (+2 more)
Nheengatu (yrl), the Amazonian Língua Geral, has 6–10 thousand speakers and is co-official in São Gabriel da Cachoeira (Brazil), yet remains poorly served by evaluated Portuguese–Nheengatu MT resources. We present the largest parallel corpus assembled for the
language (14.5k Portuguese–Nheengatu pairs), anchored by the 2023 official translation of the Brazilian Federal Constitution; the first MT evaluation set built from transcribed spontaneous
speech; and, to our knowledge, the first open-weight MT models dedicated to bidirectional Portuguese–Nheengatu translation. Three findings organize the paper. First, translation into Nheengatu on held-out constitutional articles plateaus on development data regardless of training length or model size, and much of the remaining error sits in conventional legal terms that the written sources do not attest consistently; a terminology success rate over a frozen lexicon shows that glossary mechanisms at training time and lexical constraints at decoding each raise term realization, the latter at a cost in repetitive output that chrF++ barely registers. Second, a corpus expansion that adds transcribed conversation raises
speech chrF++ by 24–27 points while leaving written domains unchanged, a register gap that written-only
benchmarks cannot see. Third, a larger NLLB model helps translation into Portuguese but not into Nheengatu on legal or
speech text. For Portuguese-to-Nheengatu translation, our final systems reach 31.7–32.8 chrF++ on article-disjoint constitutional text, 64.1–64.5 on the near-domain set (62.1 after removing items with target-side content overlap), and 45.2–47.5 on
speech.
Read the full abstract ▾
All authors Fabrício Carraro, Rodolfo Joel Zevallos, Rodrigo Gonçalves de Souza, Caio Henrique Faustino da Silva, John E. Ortega
Human-Computer Interaction
Rethinking Opinion Measurement Interfaces for Human-in-the-Loop Optimization
Changkun Ou, Sven Mayer, Daniel Buschek et al. (+1 more)
Human-in-the-loop optimization refers to systems that present choices to a human decision-maker and query their judgment to optimize system parameters iteratively. The user interface typically provides either an absolute rating scale or asks for a preferred option. However, these approaches can potentially suffer from calibration problems, inconsistencies, and completeness violations of user preferences in such a sequential evaluation. This paper revisits the design space of opinion measurement interfaces for human-in-the-loop optimization. We conducted a study (N=360) using text summarization and image color enhancement tasks to evaluate human sequential feedback in six representative opinion measurement interfaces. Based on our analysis, we recommend using a listwise approach to support sequential opinion measurement, which counterweighs individual rating and preferential choice limitations and conveys more information for subsequent use. We further discuss the trade-offs in different user interface designs and provide guidelines to inform future research.
Read the full abstract ▾
All authors Changkun Ou, Sven Mayer, Daniel Buschek, Andreas Butz
Computation and Language
Amharic-English Cross-Language Question Answering Using Statistical Machine Translation, Named Entity Recognition, and Semantic Indexing
Emebet Bekele and Fekade Getahun
The predominance of English-
language Web content creates an information-access barrier for speakers of low-resource
languages such as Amharic. This paper presents the design, imple-mentation, and evaluation of a bidirectional Amharic-English cross-
language question answering (CLQA) system that allows a user to formulate a natural-
language question in either Amharic or English and retrieve an answer from documents written in the other
language. The system integrates a statistical machine translation (SMT) component for query translation, a maximum-
entropy named entity recognition (NER) component for Amharic text processing, semantic-based indexing, question analysis, passage
retrieval, and answer extraction. A domain-diverse Amharic-English parallel corpus was used to train the SMT component, while monolingual cor-pora supported
language modeling. Documents were normalized, tokenized, stemmed, indexed, and semantically enriched before
retrieval. The original evaluation used precision and recall as the principal effectiveness measures. For Amharic-to-English
retrieval, the system achieved 72% precision and 79% recall; for English-to-Amharic
retrieval, it achieved 65% precision and 70% recall. These findings show that a modular translation-and-
retrieval architecture can provide useful cross-
language access even under low-resource conditions. This revised article preserves the original experimental claims while situating them in the context of subsequent multilingual
transformer and Amharic QA research.
Read the full abstract ▾
Artificial Intelligence
GateSight: Recurrent Memory Anticipates Attention Memory for Query-Agnostic KV Eviction in Hybrid LLMs
Junjie Li, Jiong Lou and Jie Li
Hybrid
language models interleave Gated DeltaNet (GDN) layers, whose recurrent
memory has a fixed size, with full-
attention layers, whose key-value (KV) cache grows with the context and dominates serving
memory. In this paper, we ask whether the recurrent
memory already knows what the
attention memory will need. We first show that which GDN heads of Qwen3.8-27B keep long
memories is fixed by the weights rather than by the document. Next, we define the future utility of a KV entry by how much removing it changes the
attention output over the model's own continuations, and find that the write gate, decay and write magnitude that GDN layers emit during prefill anticipate this utility before any query is seen. Because the relation is spread across many GDN heads and layers, and a linear map recovers its trend but less of its high-utility tail than a nonlinear one, we propose GateSight, which learns it with lightweight per-head multilayer perceptrons that read these signals and no
attention score. Their predicted utility decides which entries each
attention layer keeps, and how much each layer keeps is set by the model's native
attention output gate, whose mean differs by more than an order of magnitude across layers. Finally, at serving time, GateSight scores the cache once after prefill and physically releases discarded KV entries. Experimental results on SCBench and RULER-4096 with Qwen3.8-27B and, with the same recipe, the sparse-
attention model Qwen3.8-Flash-Next, up to the 262K-
token limit of the models, show that with a tenth of the
attention KV cache, GateSight is the most accurate of five query-agnostic methods on three of the four model-
benchmark pairs and leads KVzip on all four, by 9.5 to 47.2 points. Notably, it scores a full-length context with negligible overhead relative to prefill.
Read the full abstract ▾
Robotics
Beyond Fixed Visual History: Learning Motion-Adaptive Temporal Receptive Fields for Complex-Terrain Quadruped Locomotion
Hao Zhang, Yuping Yan, Yitian Hong et al. (+5 more)
Versatile quadruped locomotion depends on a tight coupling between what the
robot perceives and what its body can execute. Yet existing visual controllers typically organize depth history using fixed temporal patterns, overlooking that the same temporal span corresponds to different spatial ranges as the
robot's speed, posture, and gait phase change. Consequently, the same visual history may be insufficient in one motion state but outdated or misaligned in another. To address this limitation, we propose TempoLoco, a body-aware dynamic visual perception framework that organizes visual
memory according to the
robot's current locomotion state. TempoLoco learns multiple complementary temporal views from a long depth buffer, encodes them within a shared feature space, and dynamically aggregates them using proprioceptive history and multi-scale terrain evidence. This enables a single policy to align visual anticipation with the
robot's current motion capabilities across complex terrains, including stairs, platforms, gaps, discrete obstacles, rough ground, and slopes. From the experiment analysis, TempoLoco achieves a 78.4% success rate, compared with 58.2% for the strongest fixed-history baseline and 62.1% for uniform multi-scale fusion. In separate capability evaluations, it completes 41/45 simulated trials across nine terrain types and 23/25 real-world trials on a Unitree B2 with a fixed arm, demonstrating robust and deployable locomotion through body-aware dynamic visual perception.
Read the full abstract ▾
All authors Hao Zhang, Yuping Yan, Yitian Hong, Luying Feng, Tianhong Wang, Fengqi Dai, Wei Chen, Yaochu Jin
Machine Learning
Random Fourier Features in Near-Linear Time
Sepideh Mahabadi, Sandeep Silwal and Tal Wagner
Random
Fourier features approximate a shift-invariant kernel in $\mathbb{R}^d$, such as the Gaussian kernel $\exp(-\tfrac12\|x-y\|_2^2)$, using independent samples from its spectral measure to construct an unbiased
estimator for the kernel value with subgaussian concentration. However, generating $m$ such features requires $O(md)$ time. A long line of work on structured random features has accelerated this computation for the Gaussian kernel and other radial kernels, albeit with weaker concentration guarantees. We close this gap by showing a random feature map that can be computed in near-linear time $O((d+m)\log d)$ and yields near-subgaussian concentration. We obtain it through a new and tighter analysis of the Fastfood feature map of Le, Sarlós and Smola (ICML 2013), using Brascamp-Lieb inequalities and the Schoenberg representation of radial kernels.
Read the full abstract ▾